In the majority of excitatory pathways within the brain, namely those mediated by the neurotransmitter glutamate, the electrophysiological and molecular dissection of glutamate receptor function has led to a sophisticated understanding of how relatively correlated activity among converging developing axons can mediate strengthening and stabilization of developing inputs via the NMDA receptor (NR) and its ability to drive long term potentiation of synaptic strength (LTP). NR-LTP strengthens initial glutamatergic synapses when the activity they transmit is correlated in time. This has been established with chemical antagonists specific to the NR as well as by experimental manipulation of activity pattern in several neural preparations. However, it is not known if poorly correlated axon activity via NR-LTD weakens the many imprecisely targeted glutamate- mediated connections in developing and some mature but diseased or damaged pathways. For example, unrefined central projections are likely to be retained with uncorrected amblyopia in children, with long-term cataract obstruction of high fidelity vision or damage to the optic nerve. Whether or not abnormal NR-LTD is also involved in these conditions is hard to determine experimentally even in mouse models because antagonists of NRs block NR-LTP as well as NR-LTD. In our last NEI funding period we developed short- hairloop shRNAs against either SAP102 or PSD-95 that severely reduced or eliminated these NR synaptic scaffolds. We found that in slice recordings from single visual cortex (VC) neurons, in mice shortly after eye- opening, that SAP102KDs eliminated NR-LTD but left NR-LTP un-effected in layer 2/3 VC cells. We achieved similar results in slices from post-eye opening mouse pup VC neurons that were infected with HSV carrying different carrying different modifications that disrupted either the normal GluN2A or 2B subunit composition of the of NRs. We also have been working on a mutant mouse Flailer that lacks NR-LTD but retains NR-LTP in VC and Superior Colliculus. In this current application we propose to use a variety of anatomical, electrophysiological, and optical recording techniques, along with some modifications to the approaches described above and the Flailer mouse, that was obtained initially for unrelated experiments, to determine if, in developmenting synaptic connections young neurons lacking NR-LTD can refine the developing visual projections throughout the visual pathway in VC, Superior Colliculus and the dorsal lateral geniculate nucleus.